Stillwater-based OSU’s Center for Health Systems Innovation researchers have analyzed data from more than 18,000 patients who were hospitalized with the virus from the Cerner COVID-19 data cohort, which is a collection of de-identified and HIPAA-compliant digital health records donated to CHSI.
The research team has created two models of potential mortality risk; the first is based on patient data at the time of admission and their demographic and historical medical conditions, while the second is based on data recorded at the end of the first hospitalization day using demographics, known conditions, procedures and medications.
“The models identified a similar set of medical conditions suggested by the Centers for Disease Control and Prevention as the essential risk factors for death, such as history of diabetes, respiratory disorders and hypertension, and onset of respiratory or kidney failures, but we also found some unique ones,” said Zhuqi Miao, PhD, CHSI’s health data science program manager, according to the Sept. 2 report.
The first model can predict mortality for almost 70 percent of patient, and the second is accurate for about 75 percent of patients, Dr. Miao said.
More articles on data analytics:
Social media posts misinterpreting CDC data go viral: 4 things to know
Google releases COVID-19 search term data for 400 symptoms — will it work better than its flu tracker?
Facebook, Carnegie Mellon, U of Maryland team up on COVID-19 symptom data challenge
